TAKAHASHI AND K A N ~
1118
VOL.30
Benzimidazole N-Oxides. V. Reactions of 1,2-Dimethylbenzimidazole 3-Oxide with Acetylenecarboxylates S H I RTAKAHASHI ~ AND HIDEOK A N ~ Shionogi Research Laboratory, Shionogi and Company, La., Fukushimu-ku, Osaka, Japan Receiced November 6, 1964 The reaction of 1,2-dimethylbenzimidazole3-oxide (4) with dimethyl acetylenedicarboxylate or methyl propiolate afforded 1: 1 adducts (5 or 6 ) , respectively. The structures of these adducts were determined by spectra and chemical degradation and proved to be 1,2-dimethyl-3-(@-hydroxyviny1)benzimidazolium betaine derivatives by independent synthesis. A mechanism for this reaction, which consists of an intermediate formation of cycloadducts, has been proposed tentatively.
In previous paperslabb dealing with 1,&dipolar cycloaddition reactions of 1-methylbenzimidazole 3-oxide ( l ) , we have reported that the reaction of 1 with dimethyl acetylenedicarboxylate or methyl propiolate gives methyl methoxalyl- or methyl formyl(1-methyl2-benzimidazoly1)acetate (3a or 3b, respectively), and the formation of these products can be explained by the initial 1,&dipolar cycloaddition followed by cleavage of the isoxazoline ring of the adducts (2a and b) (Chart I).
four methyl signals (T 6.12, 6.21, 6.40, and 7.38) in 5 and three methyl signals ( T 6.26, 6.30, and 7.40) and an aldehyde proton signal (T 0.55) in 6 were observed besides the signals of benzene ring protons. In the light of these spectral data and the chemical evidence to be described below, the structures 3-(a,/3dimethoxycarbonyl-0-hydroxyviny1)- and 3-(a-methoxy carbonyl-@-hydroxyviny1)- l12-dimethylbenzimi d a zolium betaines were assigned to 5 and 6, respectively. CHARTI1
-
CHARTI
0
I
7
I
CHs 1
1
I
CHa ?a, R = C ~ C H S
b, R = H
CH3 I CH3
I CH3 6
CH* Sa, R==C&CHa
5
1HC1
b, R - H
bCI
OH
Continuing interest was fostered by these facts and our present investigation was undertaken to see if 1,2dimethylbenzimidazole 3-oxide (4) would behave as a l13-dipole on reaction with the acetylenecarboxylates. The N-oxide 4 reacted with dimethyl acetylenedicarboxylate or methyl propiolate in chloroform a t room temperature to yield a 1:l adduct (5 or 6, respectively) in excellent yield. Both of the adducts are very soluble in water and gave monohydrochlorides which in turn reproduced the free bases upon neutralization. The infrared spectra (in Nujol) of 5 and 6 in the carbonyl region showed very similar bands (1735, 1670, and 1550 cm.-l in 5 ; 1660 and 1590 cm.-* in 6) to those of 3a and 3b, respectively. The strong absorption bands near 1550 cm.-l assignable to enolate ion2 together with the positive color reaction with ferric chloride suggest the presence of an enolate group in 5 and 6. In the n.m.r. spectra (in CDC13), (1) (a) 9. Takahsshi and H. Kana, Tetrahedron Lettere. N o . 26, 1687 (1963); (b) Chem. Pharm. Bull. (Tokyo), 12, 1290 (1964). (2) R . Huisgen and H. Seidl, Tetrahedron Letters, NO.19, 2019 (1963).
I
H-C-OH
O=C
I CHz I
c1-
I
CHz I
CH3 7
9
I
CH3OH
OCHa
I
o=c-ca-
H-C-OH I
CHP
I
c1-
I
CH3
I
-C 4 H
I
I CHa
c1-
I
N I
I
CH3
CH3 10
8
CH3
APRIL1965
BENZIMIDAZOLE N-OXIDES. V CHARTIV
CHARTI11 CQH
CH20H
I
I CHi
I
1119
CHa
N
I
I
COC4CHs
CHs
CHs
@NNF CHs
13
11
17
I CHzCaCHs
I
KMnO, st60°
CO2-
16
16
HC02CH3'
I
CH2 I
14
t
CHaI
6
H
I
I CHCQCHs I CHO
a" NZ -0 CHs
@ C -:H.
18
I
CH3
aldehyde derivative) was assigned to 9. Recrystallization of 9 from methanol-acetone gave the hemiacetal (10).
The compound 8 was oxidized with potassium permanganate at 50" to a carboxylic acid (11). TreatI ment of 11 with diazomethane gave its methyl ester CHa (12), whose infrared spectrum (in Nujol) showed absorption bands at 1737 (COOCHJ and 1706 em.-' (C=O). The n.m.r. spectra (in CDC13) lacked the C-CHs signal peak observed before oxidation. By these data, the structures of 11 and 12 were deduced to bHB CHa 15 and be 3-carboxymethyl-1-methyl-2-benzimidazolinone its methyl ester, respectively (Chart 111). The structure assigned to 11 was confirmed by identiBy presuming that the electron of the enolate ion of fication with a sample prepared independently from 6 is delocalized as depicted in the figure, this assign3-(~-hydroxyethyl)-l-methyl-2-benzimidazolinone(13) ment would not be a t variance with the appearance which was obtained by treatment of l-methyl-2(3H)of an aldehyde proton signal in the n.m.r. spectrum benzimidazolinone with ethylene oxide. (Chart 11). Hydrolysis of 5 with hydrochloric acid gave a hydroWhen the permanganate oxidation of 7-10 was carchloride (7), which was converted into its free base (8) ried out a t 25", an alternative product (14) was obtained. The n.m.r. spectrum (in DzO) indicated a by neutralization. Analytical figures for 7 suggested that the two ester groups of 5 were hydrolyzed and signal peak at T 5.05 assignable to the -CH2- group in one of the resulting carboxyl groups was decarboxylated addition to those of the benzene ring protons, N-CHI and C-CHa. The presence of a carboxylate group was to yield 7. The 1i.m.r. spectra (in CFsCOOH) of 7 and 8 also indicated that disappearance of two suggested by its infrared absorption bands a t 1660 (shoulder), 1620, and 1375 cm.-l (in Kujol). These methyl groups of the esters and appearance of a - C H r results, in conjunction with the higher temperature group (signal peaks at T 4.41 and 4.00, respectively) oxidation of 14 which gave 11, indicated that 14 is 3had occurred upon hydrolysis. The infrared spectra (in Nujol) showed the presence of the -COCOOH carboxymethyl-1 ,Pdimethylbenzimidazolium betaine. group (2400-2800, 1743, and 1710 sh cm.-l) in 7 and This was confirmed by the following independent synthesis. Reaction of 1,2-dimethylbenzimidazole with the -COCOO- group (1633, 1603, and 1412 cm.-l) bromoacetic acid gave the bromide (15), which gave the in 8. On the basis of these studies it seemed most probable that the structures of 7 and 8 are represented betaine 14 upon neutralization. by 1,2-dimethyl-3-hydroxalylmethylbenzimidazolium These transformation studies support the structures chloride and its free base, a betaine. assigned to 5 and 6. To prove the structure, these Similar hydrolysis with 6 gave a crystalline product compounds were synthesized unambiguously as follows. (9),whose n.m.r. spectrum (in DzO) showed a triplet Treatment of the silver salt of 2-methylbenzimidazole at T 4.51 and a doublet at 5.50 assignable to the protons with methyl bromoacetate gave methyl (2-methyl-lof a >CH-CH2- group and indicated the disappearbenzimidazoly1)acetate (16). Reaction of 16 with diance of the methyl group of the ester 6. The infrared methyl oxalate or methyl formate in the presence of pospectrum lacked the absorption band in the carbonyl tassium methoxide gave methyl niethoxalyl(2-methyl-lregion. benzimidazoly1)acetate (17) or methyl formyl(2-methylFrom these spectral data together with elemental 1-benzimidazoly1)acetate (18), respectively (Chart IV) . analysis, the structure of 1,2-dimethyl-3-(@,@-dihyThese esters 17 and 18 reacted with methyl iodide in droxyethyl) benzimidazolium chloride (as a hydrated the presence of potassium carbonate and produced the COzH
1120
TAKAHASHI AND K A N ~ CHARTV 4
1
RC=CCO&Ha
r-0-c-R
1
VOL.30
sodium borohydride reduction of 6 did not proceed, and the material was recovered unchanged. Some of these properties, especially the abnormal aldehyde character of 6 , may be due to the contribution of enolate ion structure, which is indicated by the spectral data mentioned above.
Experimental3 L
20
I
CHs 23
\ ; L
21
5 and 6
1
CHs 22
1
desired betaines 5 and 6 , respectively. The products were proven identical by infrared comparison and mixture melting point. I n the absence of the carbonate, this reaction with 17 gave 1,2-dimethyl-3-methoxycarbonylmethylbenzimidazolium iodide (19). The same product 19 was obtained from 16 and methyl iodide. On the basis of the analogy with the reaction 1 + 3, the formation of 5 and 6 from 4 can be explained by a mechanism involving 1 ,&dipolar cycloaddition intermediates (20), although the cycloadducts could not be isolated even when the reaction was carried out a t -20". In this case, in contrast to that of the intermediate 2 for the formation of 3 from 1, the initial adducts 20 have no mobile proton a t the 2-position of the imidazoline ring and no possibility to stabilize as in the transforriiation 2 + 3 (Chart V). The rearrangement of the initial adducts to 5 and 6 may proceed via the following path. The N-0 bond breaking of the unstable isoxaeoline ring of 20 would cause the migration of the group bearing a methoxycarbonyl function from carbon to nitrogen through 1,2 shift of the Q bond (a) or through initial a-bond shift (b) followed C-C bond cleavage as illustrated in 21 and 22, respectively. An alternative mechanism involving the initial formation of a betaine (23) followed by the rearrangement to 21 through a four-membered transition state cannot be ruled out. These new type compounds showed some interesting properties. The compounds 5 and 6 are very stable to boiling niethanolic hydrochloric acid in contrast to 3a arid b which were readily hydrolyzed to methyl (1-methyl-Bbenzimidazoly1)acetate (24). On standing overnight with hydroxylamine solution in methanol. 5 and 6 were recovered unchanged, while both 3a and 3b were converted to 24.Iasb The aldehyde 6 resisted oxidation with hydrogen peroxide but was oxidized to 1,2-dimethylbenzimidazolewith potassium permanganate and did not reduce Tollens reagent. Catalytic or
1,2-DImethyl-3-(a,@-dimethoxycarbon yl-p-hydroxyvinyl)bene imidazolium Betaine ( 5 ).-Dimethyl acetylenedicarboxylate (2.20 g., 0.0155 mole) was added to a solution of 1,2-dimethylbenzimidazole 34xide (4) [colorless crystals, m.p. 178-180', prepared from the dihydrate4(3.00 g., 0.0151 mole) by azeotropic dehydration with chloroform] in chloroform (15 ml.) with stirring and cooling in an ice-water bath. Then the resulting orange solution waa allowed to stand for 0.5 hr. at room temperature. After evaporation, acetone was added to the residue to give nearly colorless crystals (4.20 g., 91%). Recrystallization from acetone gave colorless plates: m.p. 238-240' dec., ",":X: 263.5 mp (log e 4.31) and 277.2 (4.16). Anal. Calcd. for C16HlsNzOs: c, 59.20; H , 5.30; N, 9.21. Found: C, 59.29; H , 5.55; N , 9.04. The same product was obtained when the dicarboxylic eater was added a t -20' and the solution was allowed to stand a t this temperature for 5 min., then evaporated under -20'. This product waa recrystallized from water to give the hydrate of 5 aa white prisms, m.p. ca. 210-240'. Anal. Ctalcd. for C16Hl,NzOa~H~O: C, 55.89; H, 5.63; N, 8.69; HzO, 5.59. Found: C, 55.78; H, 5.74; N, 8.69; HnO, 5.18. 1,Z-Dimethyl-J-(a-methoxycarbonyl-8-hydroxyPiny1)benzimidazolium Betaine (6).-This compound waa obtained from 4 (dihydrate, 3.00 g., 0.0151 mole) and methyl propioltate (1.30 g., 0.0155 mole) by the same procedure as for the synthesis of 5 ; yield 3.60 g. (96%). Recrystallization from dichloromethaneether gave colorless prisms: m.p. 212' dec., X",l"." 256.7 mp (log c 4.42) and 277.0 (4.07). Anal. Calcd. for ClaHliNzOa: C, 63.40; H , 5.73; N, 11.38. Found: C, 63.58; H, 6.02; N, 11.64. Hydrolysis of 5.-A mixture of 5 (1.00 g.) in 6 N hydrochloric acid (10.0 ml.) waa refluxed for 5 hr. The resulting colorless solution waa evaporated and the crystalline residue waa recrystallized from water-acetone to give 1,2dimethy1-3-hydroxalylmethylbenzimidaiolium chloride (7) as colorless prisms (0.90 g., 95%): m.p. 170' dec.; 255.0 mp (log c 3.80), 263.5 (3.82), 270.5 (3.93), and 277.6 (3.96). Anal. Calcd. for ClzH1~CINzOa.H~O:C, 50.26; H , 5.27; N, 9.77. Found: C, 50.29; H, 5.40; N,9.56. 1,2-Dimethyl-3-hydroxalyhethylbenzimidazoliumBetaine ( 8 ) - A solution of 7 in water waa adjusted to pH 3-4 by aqueous sodium hydrogencarbonate solution, then evaporated. The residue waa extracted with absolute ethanol and the ethanol was evaporated. Recrystallization from water-acetone gave colorless prisms, m.p. 170" dec. Anal. Calcd. for Ci2HlzNzOa'HzO: C, 57.59; H, 5.64; N , 11.20; H ~ 0 , 7 . 2 0 . Found: C, 57.65; H,5.96; N , 11.01; HzO, 7 62. Hydrolysis of 6.-A solution of '6 (12 0 g.) in 6 N hydrochloric acid (12 ml.) WRS refluxed for 7 hr., then evaporated. The colorless residue waa recrystallized from water-acetone to give 1,Z-dimethyl-3-(p,@dihydroxyethyl)benzimidazolium chloride (9) aa colorless prisms or plates (1.00 g., 85%): m.p. 210' dec. 255.0 mp (log t 3.76), 263.5 (turning brown above 150'); (3.79), 270.3 (3.91), and 277.3 (3.95). Anal. Calcd. for C11H16ClN~02: C, 54.43; H, 6.23; N, 11.54. Found: C, 54.27; H, 6.49; N, 11.42. (3) All melting points were taken on a Kofler hot stage and are uncorrected. Solvents were removed under reduced pressure. Each identification WBB made by comparison of the infrared spectrum, and, if the sample had a melting point, it was also compared by mixed fusion. Infrared spectra were recorded on a K6ken infrared spectrophotometer, Model IR-9. N.m.r. spectra were obtained on a Varian A-60 analytical n.m.r. apectrometer in CDCla or CFICOOH containing tetramethylsilane as an internal reference or in DIO containing dioxane. Unless otherwise stated, each signal pattern is singlet. (4) 9. Takahaehi and H. KanG, Cham. Phorm. BUZZ. (Tokyo), 11, 1375 (1863).
APRIL1965
BENZIMIDAZOLE N-OXIDES. V
When the colorless residue waa recrystallized from methanolacetone, the methyl acetal 10 was obtained as colorless prisms, m.p. 180" dec. Anal. Calcd. for C12HlrClN202: C, 56.14; H, 6.67; N, 10.91. Found: C,55.93; H,6.99; N , 10.85. Oxidation of 8 to 11.-To a solution of 8 (0.40 g.) in water (20 ml.) was added finely ground potassium permanganate with stirring a t 50' until the color of the oxidizing agent lasted for a time (ca. 0.9 g.). After removal of manganese dioxide by filtration, the filtrate waa acidified with 6 N hydrochloric acid to give 3-carboxymethyl-l-methyl-2-benzimidazolino1~e ( 11) aa colorless crystals (0.18 g., 55%). Recrystallization from water gave colorless prisms: m.p. 218-219'; 230.4 mp (log c 3.90), 282.0 (3.95), and 287.0 (3.89 sh). Anal. Calcd. for CloHloNzOj: C, 58.25; H , 4.89; N , 13.58. Found: C, 58.27; H, 5.07; N , 13.60. 3-Methoxycarbonylmethyl-l-methyl-2-benzimidazolinone (12).-To a suspension of 11 (158 mg.) in chloroform (5.0 ml.) waa added a solution of diazomethane in ether. The starting material dissolved rapidly into the solution as the reaction proceeded. ReInoval of the solvent left a colorless oil, which solidified on cooling. The product waa purified by chromatography on alumina with chloroform, then recrystallized from carbon tetrachloride-petroleum ether (b.p. 60-70') to give colorless plates (98 mg.), m.p. 121-123'. Anal. Calcd. for CllH12N20j: C , 59.99; H , 5.49; N, 12.72. Found: C, 60.29; H, 5.57; N, 12.51. 3-Carboxymethyl-l,2-dimethylbenzimidazolium Betaine (14). -To a solution of 7 (1.00 g.) in water (30ml.) waaadded finely ground potassium permanganate with stirring a t 25' until the color of the oxidizing agent lasted for a time (ca. 0.55 g.). After removal of manganese dioxide by filtration, the filtrate waa evaporated and the residue was extracted with absolute ethanol. After the solvent waa removed, recrystallization of the residue from ethanol-acetone gave colorless prisms (0.66 g., 827'): : 255.5 mp (log c 3.57), 264.0 (3.60), 271.0 m.p. >230"; XFH (3.71), and 278.0 (3.74). Anal. Calcd. for C11H12N20z*1.5€120: C, 57.13; H, 6.54; N , 12.12; HzO, 11.70. Found: C, 57.14; H, 6.63; N, 12.07; H20, 12.15. This compound was obtained by the oxidation of 8, 9, or 10 by the same procedure. Oxidation of 14 to 11.-Oxidation of 14 (0.20 9.) by the same procedure as for 8 gave 11 (0.08 g.), which waa identified with the sample obtained above. 3-(~-Hydroxyethyl)-l-methyl-2-benzimidazolinone ( 13).-To a solution of ethylene oxide (1.0 ml.) in water (5.0 ml.) were added l-methyl-2(3H)-benzimidazolinone (0.50 9.) and aqueous sodium hydroxide solution (lOyG,1 drop). After having been allowed to stand for 3 days with occasional shaking, the resulting solution was extracted with three 5-ml. portions of chloroform. Removal of the solvent gave a colorless oil (0.7 g,), chromatography of which on alumina with ethanol gave colorless crystals (0.30 g., 46y0). Recrystallization from carbon tetrachloride gave colorless prisms, m.p. 108-109'. Anal. Calcd. for C10H12N202: C, 62.48; H, 6.29; N, 14.58. Found: C, 62.61; H, 6.35; N , 14.40. Oxidation of 13 to 11.-Oxidation of 13 (0.20 9.) by the same procedure as for 8 gave 11 (0.08 g.). This product and ita methyl ester were identical with the samples obtained from 8, respectively 3-Carboxymethyl-1,2-dimethylbenzimidazolium Bromide (15).-A solution of 1,2-dimethylbenzimidazoles(0.15 9.) and bromoacetic acid (0.15 g.) in ethanol (5.0 ml.) waa heated at 120" in a sealed tube for 10 hr., then evaporated. The residue waa recrystallized from ethanol-acetone to give colorless small needlea (O.O7g., 23%), m.p. ca. 170'. Anal. Calcd. for CllHlaBrN2O2~0.5Hz0:C, 44.89; H, 4.80; N , 9.53. Found: C, 44.84; H, 4.88; N , 9.41. The bromide waa dissolved in water and made alkaline with potassium carbonate. After evaporation, the residue waa extracted with absolute ethanol. Acetone waa added to the concentrated ethanolic solution to give colorless prisms, m.p. >230'. This compound waa identical with above-obtained 14. Methyl (2-Methyl-1-benzimidazoly1)acetate(16).-To a solution of silver nitrate (8.50 g., 0.050 mole) in water (15 ml.) were added ethanol (50 ml.), then aqueous ammonia (30%) until the first precipitate dissolved (ca. 10 ml.). The resulting solution was added to a solution of 2-methylbenzimidazole (6.60 g., ( 5 ) 0. Fischer, Bev., 91, 2838 (1892).
1121
0.050 mole) in ethanol (50 ml.) to give a white precipitate, which waa collected by filtration (11.5 g.). To a suspension of the thus obtained silver salt of 2-methylbenzimidazole (9.55 g., 0.040 mole, finely ground) in toluene (75 ml.) w a ~added methyl bromoacetate (6.30 g., 0.041 mole) dropwise with stirring and heating under reflux and, after the addition, stirring and heating were continued for 5 hr. The resulting solution waa separated by decantation from the tarry residue and evaporated to give brown tar, which was chromatographed on alumina with chloroform to give a crystalline product (3.20 g., 37%). Recrystallization from benzene (plates), then isopropyl ether (needles) or carbon tetrachloride (prisms), gave colorless crystals: m.p. 117-119", ~2::"' 1735 cm.-' ( C = O ) . 9 n d . Calcd. for C ~ ~ H I Z N Z C, O ~64.69; : H, 5.92; N , 13.72. Found: C, 64.73; H , 6.08; N, 13.93. Methyl Methoxalyl( 2-methyl-1-benzimidazoly1)acetate (17) .To a suspension of potsssium methoxide in absolute ether, prepared by addition of methanol (0.50 ml.) to a suspension of powdered potassium (0.15 g., 0.00375 mole) in absolute ether (10 ml.), were added 16 (0.61 g., 0.0030 mole) and dimethyl oxalate (0.35 g., 0.0030 mole) with stirring a t room temperature. After stirring for 3 hr., the mixture waa allowed to stand overnight a t room temperature. The resulting crystalline precipitate wm collected by filtration, waahed with ether, and dissolved in a small amount of water. The solution waa acidified (pH 4) with 6 N hydrochloric acid to give colorless crystals (0.55 g., 63y0), Recrystallization from methanol-acetone (or chloroform-benzene) gave colorless plates: m.p. 212' dec.; 1733 263.0 mp (log e 4.34), 266.8 (4.33 sh), and 275.6 (4.21); '":Y and 1682 (C==O), and 1534cm.-' (C=C-0-); n.m.r. (inCDCb) T 2.40-2.75 (benzene ring, complex pattern), 6.07 (COOCH,), 6.34 (COOCH,), and 7.62 (C-CHa). A n d . Calcd. for CIIHIINZO~:C, 57.93; H, 4.86; N, 9.65. Found: C, 57.88; H, 5.06; N, 9.56. Methyl Formyl( 2-methyl-1-benzimidazoly1)acetate(18) .-To a suspension of potassium methoxide in absolute ether, prepared by addition of methanol (1.00 ml.) to a suspension of powdered potassium (0.24 g., 0.0060 mole) in absolute ether (20 ml.), were added methyl formate (0.60 g., 0.0100 mole) then 16 (1.02 g., 0.0050 mole) with stirring at room temperature. After stirring for 3 hr., the mixture waa allowed to stand overnight at room temperature. The resulting gummy product waa waahed with ether after separation from the solution by decantstion, then dieaolved in a small amount of water. The solution waa acidified (pH 4) with 6 N hydrochloric acid and extracted with chloroform. Removal of the solvent gave a colorless oil (0.80 g., 69%), which solidified on standing. Recrystallization from methanol-acetone (or chloroform-benzene) gave colorless plates: m.p. 207' dec.; " : :A 257.5 mp (log L 4.36), 267.2 (4.26 ah), and 275.7 (4.08); ~::Y 1677 ( C 4 ) and 1564 cm.-l (C=C-0-, broad); n.m.r. (in CDCls) T 1.13 (CHO), 2.60-2.80 (benzene ring, complex pattern), 6.29 (COOCHJ), and 7.62 (C-CHj). Anal. Calcd. for C12H12NzOj: C, 62.02; H, 5.21; N, 12.06. Found: C,61.80; H,5.41; N, 12.12. Synthesis of 5 from 17.--To a solution of 17 (145 mg., 0.50 mmole) in methanol (1.0 ml.) were added finely ground potassium carbonate (350 mg., large excess) and methyl iodide (0.10 ml. excess), and the mixture waa heated in tt sealed tube a t 110' for 2 hr. After removal of insoluble materials by filtration, the filtrate was evaporated. The residue waa extracted with chloroform, and removal of the solvent and addition of acetone then gave colorless crystals (105 mg., 69y0), m.p. 238-240' dec., which were identical with 5 obtained from 4 and dimethyl acetylenedicarboxylate. Synthesis of 6 from 18.-This experiment waa carried out by the same procedure mentioned above. From 18 (232 mg., 1.0 mmole), potassium carbonate (350mg.), methyl iodide (0.12 ml., 3.0 mmolea), and methanol (1.0 ml.), 6 (150 mg., 610/,), ,m.p. 212" dec., waa obtained. This compound was identical with 6 obtained from 4 and methyl propiolate. 1,2-Dimethyl-3-methoxycarbonylmethylbenzimidazolium Iodide (19). A. From 16.-To a solution of 16 (204 mg., 1.0 mmole) in methanol (2.0 ml.) wm added methyl iodide (0.075 ml., 1.2 mmolea), and the solution was heated in a sealed tube a t 120" for 5 hr. After evaporation of the resulting slightly brown solution, acetone was added to the residue to give colorless crystals (0.20 g.). Recrystallization from methanol-isopropyl ether gave colorless plates, m.p. 195-197' dec.
1122
WATSONAND HAAKE
VOL.30
Anal. Calcd. for CIZHISIN~OZ: C, 41.64; H , 4.37; N, 8.09. Oxidation of 6 with Potassium Permanganate.-To a solution Found: C, 41.96; H,4.64; N , 8.09. of 6 (0.40 g.) in water (6.0 ml.) waa added finely ground potasB. From 17.-A solution of 17 (145 mg., 0.50 mmole) and sium permanganate with stirring a t 25', until the color of the methyl iodide (0.20 ml., excess) in methanol (3.0 ml.) waa heated oxidizing agent lasted for a time. Crystals separated as the in a sealed tube a t 120" for 2.5 hr., then evaporated. Recrystalreaction proceeded. At the end of the reaction, the mixture lization of the residue from methanol-isopropyl ether gave colorwaa heated to dissolve the crystalline product and filtered to less plates (0.12 g.), m.p. 195-197" dec. remove manganese dioxide. After cooling, the product, (0.20 This compound was identical with the above-obtained one. g., m.p. 112-114") waa collected by filtration; it was identical 1,2-Dimethyl-3-(~,~-dimethoxycarbonyl-~-hydro~inyl)benziwith authentic 1,2-dimethylben~imidazole.~ midazolium Chloride.-To a solution of 5 (0.40 9.) in water Attempted Reduction of 6 with Sodium Borohydride.-Sodium (5.0 ml.) was added 6 N hydrochloric acid (1.0 ml.), and the borohydride (50 mg.) was added to a solution of 6 (200 mg.) in resulting solution was evaporated. The crystalline residue was water (5.0 ml.), and the solution was heated at 60' for 2 hr. recrystallized from methanol-isopropyl ether to give colorless The starting material (180 mg.) was recovered. plates (0.35 g.), m.p. 130" dec. Attempted Reduction of 6 with Adams Catalyst.-A mixture Anal. Calcd. for C1~H17C1NZO6~Hz0: C, 50.21; H , 5.33; of 6 (246 mg.) in methanol (20 ml.) containing platinum catalyst N , 7 . 8 1 . Found: C,50.35; H, 5.66; N,8.09. (prepared from A d a m catalyst, 50 mg.) was treated with hydroThe aqueous solution,of this compound waa neutralized with gen a t ordinary temperature and pressure. No hydrogen was aqueous sodium hydrogen carbonate solution then evaporated. absorbed during 6 hr. The starting material (240 mg.) was reThe residue was extracted with chloroform, and the solvent covered. was evaporated to give 5, m.p. 238-240" dec., which was identiReaction of Methyl Methoxalyl( 1-methyl-2-benzimidazoly1)cal with above-obtained one. acetate (3a) with Methanolic Hydrogen Chloride.-To a solution 1,2-Dimethyl-3-( ~-methoxycarbonyl-j3-hydroxyvinyl)benzimid- of 3a (0.40 g.) in methanol (3.0 ml.) waa added methanolic azolium Chloride.-This compound was obtained from 6 by the hydrogen chloride (lo%, 1.0 ml.) under cooling in a cold-water same procedure as mentioned above. Recrystallization from bath. After evaporation, acetone was added to the residue to methanol-isopropyl ether gave colorless plates, m.p. 188-190" give colorless crystals (0.30 g.). Recrystallization from methyl dec acetate gave methyl (1-methyl-2-benzimidazo1yl)acetate hydroAnal. Calcd. for C1SH16ClN203: C, 55.23; H, 5.35; N , 9.91. chloride as colorless prisms, m.p. 147-149". Found: C, 55.14; H , 5.42; N, 10.03. A n d . Calcd. for CIIHIZNZO~.HC~: C, 54.89; H, 5.45; N, Neutralization of this compound by the same procedure as 11.64. Found: C,54.79; H , 5 . 7 0 ; N, 11.66. . mentioned above reproduced 6 . Neutralization of this compound gave methyl ( l-methyl-2Treatment of 5 with Methanolic Hydrogen Chloride .-A solubenzimidazoly1)acetate (24), m .p. 66-68" tion of 5 (0.30 g.) in methanolic hydrogen chloride (lo%,3.0 ml.) Reaction of Methyl Formyl( 1-methyl-2-benzimidazoly1)acetate was refluxed for 10 hr., then evaporated. After neutralization (3b)with Methanolic Hydrogen Chloride .-This experiment was of the residue, the starting material (0.27 9.) waa recovered. carried out by the same procedure as mentioned above. By Treatment of 6 with Methanolic Hydrogen Chloride.-This this treatment, 3b also gave methyl (1-met~hyl-2-benzimidazolyl)experiment waa carried out by the same procedure aa mentioned acetate hydrochloride, m.p. 147-149', which was identical with above. The same result, recovery of the starting material, the above obtained one. was obtained by this treatment. Acknowledgment.-The authors express their gratiTreatment of 5 with Hydroxylamine.-To a solution of hydroxylamine hydrochloride (140 mg., 2.0 mmoles) and sodium tude to Professor Emeritus E. Ochiai of the University acetate (200 mg., 2.4 mmoles) in water (1.0 ml.) and methanol of Tokyo and Dr. K. Takeda, Director of this labora(4.0 ml.) was added 5 (304 mg., 1.0 mmole). The resulting tory, for their helpful guidance and encouragement, and solution waa allowed to stand a t room temperature for 3 days. their appreciation to Drs. Y. Matsui and K. Tori for After evaporation, the starting material (280 mg.) waa recovered. their helpful discussion on the infrared and n.m.r. Treatment of 6 with Hydroxylamine.-This experiment was carried out by the same procedure as mentioned above. The spectra. Thanks are also due to Messrs. I. Tanaka starting material waa recovered unchanged by this treatment. and K. Aono for ultraviolet and n.m.r. spectral measureAttempted Oxidation of 6 with Hydrogen Peroxide.-To a ments, to the members of the analysis room of this solution of 6 (0.60 8.) in methanol (10.0 ml.) was added hydrogen laboratory for elemental analysis, and to Mr. S. Hashiperoxide (300j0, 2.0 ml.) and the solution was refluxed for 2 hr. mot0 for his technical assistance. After evaporation, the starting material (0.50 9.) was recovered.
.
Decarboxylation of Oxamic and Oxanilic Acids in Aniline' JOSEPHWATSONAND PAULHAAKE Department of Chemistry, University of CalifonzzaJ2 Los Angeks, California 900.94 Received August BO, 1964 The decarboxylation of oxamic acid in aniline is complicated by a series of concurrent reactions and is not reversible as previously reported, but is accompanied by amidation and transamidation producing as final products oxanilide and formanilide.
As part of his extensive research on rates of decarboxylations in a variety of solvents, L. W. Clark has studied the decarboxylation of oxamic acid (I) in aniline,3 o-t~luidine,~ quinolineJ4dimethyl sulfoxide,6 and triethyl p h o ~ p h a t e . ~In the paper on the decarboxylation of oxamic acid (eq. 1) in aniline and o-toluidine as (1) The research on which this work is based was supported in part by Grants OM-9294 and AM-6870 from the U. 9. Public Health Service. (2) Contribution No. 1789 from the Department of Chemistry. (3) L. W. Clark, J. P h ~ s Chem., . 65, 180 (1961). (4) L. W. Clark, abzd., 65, 659 (1961). (5) L. W . Clark, ibzd., 65, 1651 (1961).
solvents,* there are several anomalous results including unusually high and positive AS* values (Table I) and the formation of a white precipitate (A) in the condenser concurrent with resorption of the gaseous COz produced during the decarboxylation, Clark attribut,ed this to reversal of the decarboxylation with formation of oxamic acid in the ~ o n d e n s e r . ~In a later paperj4it was
